Reproduce “Modeling Subjective Urban Perception with Human Gaze” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Lin Che, Xi Wang, Marc Pollefeys, Konrad Schindler, Martin Raubal, Peter Kiefer
Researchers at ETH Zurich developed a framework that incorporates human gaze patterns to model subjective urban perception. Integrating gaze dynamics with visual scene representations consistently enhances the prediction of how individuals perceive urban environments (e.g., safety, wealth, boredom) compared to models relying only on static image content.
Key insights
- Human gaze patterns, including fixation location, duration, and subsequent saccade length, carry significant subject-specific predictive signals for subjective urban perception.
- Integrating gaze dynamics with visual scene representations consistently improves urban perception prediction, offering complementary subject-specific cues beyond what is extractable from visual content alone.
- The correct spatial and temporal correspondence between gaze behavior and scene elements (semantic areas or visual patches) is crucial for achieving enhanced prediction performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.